Release 0.5.0 (#1024)

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: himself65 <himself65@users.noreply.github.com>

.github/workflows/lint_on_push_or_pull.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

.github/workflows/preview.yml

@@ -0,0 +1,28 @@
+name: Publish Preview
+on: [pull_request]
+
+jobs:
+  pre_release:
+    name: Pre Release
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout Repo
+        uses: actions/checkout@v4
+
+      - uses: pnpm/action-setup@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: "pnpm"
+
+      - name: Install dependencies
+        run: pnpm install
+
+      - name: Build
+        run: pnpm run build
+
+      - name: Pre Release
+        run: pnpx pkg-pr-new publish ./packages/*

.github/workflows/publish_github.yml

@@ -12,7 +12,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/release.yml

@@ -15,7 +15,7 @@ jobs:
       - name: Checkout Repo
         uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/test.yml

@@ -23,7 +23,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
 
       - name: Setup Node.js
         uses: actions/setup-node@v4
@@ -45,7 +45,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -60,7 +60,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -97,7 +97,7 @@ jobs:
     name: Build LlamaIndex Example (${{ matrix.packages }})
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
@@ -116,7 +116,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: pnpm/action-setup@v3
+      - uses: pnpm/action-setup@v4
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:

README.md

@@ -76,7 +76,7 @@ main();
 node --import tsx ./main.ts
 ```
 
-### Next.js
+### React Server Component (Next.js, Waku, Redwood.JS...)
 
 First, you will need to add a llamaindex plugin to your Next.js project.
 
@@ -154,40 +154,6 @@ export async function chatWithAgent(
 }
 ```
 
-### Cloudflare Workers
-
-```ts
-// src/index.ts
-export default {
-  async fetch(
-    request: Request,
-    env: Env,
-    ctx: ExecutionContext,
-  ): Promise<Response> {
-    const { setEnvs } = await import("@llamaindex/env");
-    // set environment variables so that the OpenAIAgent can use them
-    setEnvs(env);
-    const { OpenAIAgent } = await import("llamaindex");
-    const agent = new OpenAIAgent({
-      tools: [],
-    });
-    const responseStream = await agent.chat({
-      stream: true,
-      message: "Hello? What is the weather today?",
-    });
-    const textEncoder = new TextEncoder();
-    const response = responseStream.pipeThrough(
-      new TransformStream({
-        transform: (chunk, controller) => {
-          controller.enqueue(textEncoder.encode(chunk.response.delta));
-        },
-      }),
-    );
-    return new Response(response);
-  },
-};
-```
-
 ## Playground
 
 Check out our NextJS playground at https://llama-playground.vercel.app/. The source is available at https://github.com/run-llama/ts-playground

apps/docs/CHANGELOG.md

@@ -1,5 +1,74 @@
 # docs
 
+## 0.0.41
+
+### Patch Changes
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/examples@0.0.7
+
+## 0.0.40
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.39
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.38
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.37
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.36
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.35
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.34
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.33
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.32
 
 ### Patch Changes

apps/docs/docs/getting_started/concepts.md

@@ -62,7 +62,7 @@ These building blocks can be customized to reflect ranking preferences, as well
 
 [**Retrievers**](../modules/retriever.md):
 A retriever defines how to efficiently retrieve relevant context from a knowledge base (i.e. index) when given a query.
-The specific retrieval logic differs for difference indices, the most popular being dense retrieval against a vector index.
+The specific retrieval logic differs for different indices, the most popular being dense retrieval against a vector index.
 
 [**Response Synthesizers**](../modules/response_synthesizer.md):
 A response synthesizer generates a response from an LLM, using a user query and a given set of retrieved text chunks.

apps/docs/docs/modules/data_loaders/llama_parse/index.mdx

@@ -27,23 +27,25 @@ They can be divided into two groups.
 
 - `apiKey` is required. Can be set as an environment variable `LLAMA_CLOUD_API_KEY`
 - `checkInterval` is the interval in seconds to check if the parsing is done. Default is `1`.
-- `maxTimeout` is the maximum timout to wait for parsing to finish. Default is `2000`
+- `maxTimeout` is the maximum timeout to wait for parsing to finish. Default is `2000`
 - `verbose` shows progress of the parsing. Default is `true`
 - `ignoreErrors` set to false to get errors while parsing. Default is `true` and returns an empty array on error.
 
 #### Advanced params:
 
 - `resultType` can be set to `markdown`, `text` or `json`. Defaults to `text`. More information about `json` mode on the next pages.
-- `language` primarly helps with OCR recognition. Defaults to `en`. Click [here](../../../api/type-aliases/Language.md) for a list of supported languages.
+- `language` primarily helps with OCR recognition. Defaults to `en`. Click [here](../../../api/type-aliases/Language.md) for a list of supported languages.
 - `parsingInstructions?` Optional. Can help with complicated document structures. See this [LlamaIndex Blog Post](https://www.llamaindex.ai/blog/launching-the-first-genai-native-document-parsing-platform) for an example.
 - `skipDiagonalText?` Optional. Set to true to ignore diagonal text. (Text that is not rotated 0, 90, 180 or 270 degrees)
 - `invalidateCache?` Optional. Set to true to ignore the LlamaCloud cache. All document are kept in cache for 48hours after the job was completed to avoid processing the same document twice. Can be useful for testing when trying to re-parse the same document with, e.g. different `parsingInstructions`.
 - `doNotCache?` Optional. Set to true to not cache the document.
 - `fastMode?` Optional. Set to true to use the fast mode. This mode will skip OCR of images, and table/heading reconstruction. Note: Non-compatible with `gpt4oMode`.
-- `doNotUnrollColumns?` Optional. Set to true to keep the text according to document layout. Reduce reconstruction accuracy, and LLM's/embedings performances in most cases.
-- `pageSeperator?` Optional. The page seperator to use. Defaults is `\\n---\\n`.
+- `doNotUnrollColumns?` Optional. Set to true to keep the text according to document layout. Reduce reconstruction accuracy, and LLMs/embeddings performances in most cases.
+- `pageSeparator?` Optional. The page separator to use. Defaults is `\\n---\\n`.
 - `gpt4oMode` set to true to use GPT-4o to extract content. Default is `false`.
 - `gpt4oApiKey?` Optional. Set the GPT-4o API key. Lowers the cost of parsing by using your own API key. Your OpenAI account will be charged. Can also be set in the environment variable `LLAMA_CLOUD_GPT4O_API_KEY`.
+- `boundingBox?` Optional. Specify an area of the document to parse. Expects the bounding box margins as a string in clockwise order, e.g. `boundingBox = "0.1,0,0,0"` to not parse the top 10% of the document.
+- `targetPages?` Optional. Specify which pages to parse by specifying them as a comma-separated list. First page is `0`.
 - `numWorkers` as in the python version, is set in `SimpleDirectoryReader`. Default is 1.
 
 ### LlamaParse with SimpleDirectoryReader

apps/docs/docs/modules/data_loaders/llama_parse/json_mode.mdx

@@ -8,7 +8,7 @@ In JSON mode, LlamaParse will return a data structure representing the parsed ob
 
 ## Usage
 
-For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method. Currently it can't be used with `SimpleDirectoryReader`.
+For Json mode, you need to use `loadJson`. The `resultType` is automatically set with this method.
 More information about indexing the results on the next page.
 
 ```ts

apps/docs/docs/modules/llms/available_llms/bedrock.md

@@ -15,7 +15,7 @@ Settings.llm = new Bedrock({
 });
 ```
 
-Currently only supports Anthropic models:
+Currently only supports Anthropic and Meta models:
 
 ```ts
 ANTHROPIC_CLAUDE_INSTANT_1 = "anthropic.claude-instant-v1";
@@ -25,6 +25,10 @@ ANTHROPIC_CLAUDE_3_SONNET = "anthropic.claude-3-sonnet-20240229-v1:0";
 ANTHROPIC_CLAUDE_3_HAIKU = "anthropic.claude-3-haiku-20240307-v1:0";
 ANTHROPIC_CLAUDE_3_OPUS = "anthropic.claude-3-opus-20240229-v1:0"; // available on us-west-2
 ANTHROPIC_CLAUDE_3_5_SONNET = "anthropic.claude-3-5-sonnet-20240620-v1:0";
+META_LLAMA2_13B_CHAT = "meta.llama2-13b-chat-v1";
+META_LLAMA2_70B_CHAT = "meta.llama2-70b-chat-v1";
+META_LLAMA3_8B_INSTRUCT = "meta.llama3-8b-instruct-v1:0";
+META_LLAMA3_70B_INSTRUCT = "meta.llama3-70b-instruct-v1:0";
 ```
 
 Sonnet, Haiku and Opus are multimodal, image_url only supports base64 data url format, e.g. `data:image/jpeg;base64,SGVsbG8sIFdvcmxkIQ==`

apps/docs/package.json

@@ -1,6 +1,6 @@
 {
   "name": "docs",
-  "version": "0.0.32",
+  "version": "0.0.41",
   "private": true,
   "scripts": {
     "docusaurus": "docusaurus",

examples/CHANGELOG.md

@@ -1,5 +1,15 @@
 # examples
 
+## 0.0.7
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/core@0.1.0
+
 ## 0.0.6
 
 ### Patch Changes

examples/README.md

@@ -9,7 +9,7 @@ make sure you have basic knowledge of the [LlamaIndexTS](https://ts.llamaindex.a
 # export your API key
 export OPENAI_API_KEY="sk-..."
 
-npx ts-node ./chatEngine.ts
+npx tsx ./chatEngine.ts
 ```
 
 ## Build your own RAG app

examples/agent/azure_dynamic_session.ts

@@ -0,0 +1,54 @@
+import "dotenv/config";
+
+import {
+  DefaultAzureCredential,
+  getBearerTokenProvider,
+} from "@azure/identity";
+import { AzureDynamicSessionTool, OpenAI, ReActAgent } from "llamaindex";
+
+async function main() {
+  const credential = new DefaultAzureCredential();
+  const azureADTokenProvider = getBearerTokenProvider(
+    credential,
+    "https://cognitiveservices.azure.com/.default",
+  );
+
+  const azure = {
+    azureADTokenProvider,
+    deployment: process.env.AZURE_OPENAI_DEPLOYMENT ?? "gpt-35-turbo",
+  };
+
+  // configure LLM model
+  const llm = new OpenAI({
+    azure,
+  });
+
+  const azureDynamicSession = new AzureDynamicSessionTool();
+
+  // Create an ReActAgent with the azure dynamic session tool
+  const agent = new ReActAgent({
+    llm,
+    tools: [azureDynamicSession],
+    // verbose: true,
+    systemPrompt: `You are a Python interpreter.
+      - You are given tasks to complete and you run python code to solve them.
+      - The python code runs by the python runtime. Every time you call $(interpreter) tool, the python code is executed in a separate cell. It's okay to make multiple calls to $(interpreter).
+      - You can run any python code you want in a secure environment.
+      - For images, return the full URL, not the base64 data.
+      - Return any image content as an HTML tag with the src attribute set to the URL of the image.`,
+  });
+
+  // Chat with the agent
+  const response = await agent.chat({
+    message:
+      "plot a chart of 5 random numbers and save it to /mnt/data/chart.png",
+    stream: false,
+  });
+
+  // Print the response
+  console.log({ response });
+}
+
+void main().then(() => {
+  console.log("Done");
+});

examples/anthropic/agent.ts

@@ -2,7 +2,7 @@ import { Anthropic, FunctionTool, Settings, WikipediaTool } from "llamaindex";
 import { AnthropicAgent } from "llamaindex/agent/anthropic";
 
 Settings.callbackManager.on("llm-tool-call", (event) => {
-  console.log("llm-tool-call", event.detail.payload.toolCall);
+  console.log("llm-tool-call", event.detail.toolCall);
 });
 
 const anthropic = new Anthropic({

examples/astradb/README.md

@@ -24,7 +24,7 @@ Here are two sample scripts which work well with the sample data in the Astra Po
 
 Loads and queries a simple vectorstore with some documents about Astra DB
 
-run `ts-node astradb/example`
+run `tsx astradb/example`
 
 ## Movie Reviews Example
 
@@ -32,10 +32,10 @@ run `ts-node astradb/example`
 
 This sample loads the same dataset of movie reviews as the Astra Portal sample dataset. (Feel free to load the data in your the Astra Data Explorer to compare)
 
-run `npx ts-node astradb/load`
+run `npx tsx astradb/load`
 
 ### Use RAG to Query the data
 
 Check out your data in the Astra Data Explorer and change the sample query as you see fit.
 
-run `npx ts-node astradb/query`
+run `npx tsx astradb/query`

examples/chromadb/README.md

@@ -6,7 +6,7 @@ Export your OpenAI API Key using `export OPEN_API_KEY=insert your api key here`
 
 If you haven't installed chromadb, run `pip install chromadb`. Start the server using `chroma run`.
 
-Now, open a new terminal window and inside `examples`, run `pnpm dlx ts-node chromadb/test.ts`.
+Now, open a new terminal window and inside `examples`, run `pnpm dlx tsx chromadb/test.ts`.
 
 Here's the output for the input query `Tell me about Godfrey Cheshire's rating of La Sapienza.`:
 

examples/cloud/README.md

@@ -21,7 +21,7 @@ export LLAMA_CLOUD_BASE_URL="https://api.staging.llamaindex.ai"
 This example is using the managed index named `test` from the project `default` to create a chat engine.
 
 ```shell
-pnpx ts-node cloud/chat.ts
+pnpx tsx cloud/chat.ts
 ```
 
 ## Query Engine
@@ -29,7 +29,7 @@ pnpx ts-node cloud/chat.ts
 This example shows how to use the managed index with a query engine.
 
 ```shell
-pnpx ts-node cloud/query.ts
+pnpx tsx cloud/query.ts
 ```
 
 ## Pipeline
@@ -37,5 +37,5 @@ pnpx ts-node cloud/query.ts
 This example shows how to create a managed index with a pipeline.
 
 ```shell
-pnpx ts-node cloud/pipeline.ts
+pnpx tsx cloud/pipeline.ts
 ```

examples/cloud/chat.ts

@@ -6,7 +6,7 @@ import { ContextChatEngine, LlamaCloudIndex } from "llamaindex";
 async function main() {
   const index = new LlamaCloudIndex({
     name: "test",
-    projectName: "default",
+    projectName: "Default",
     baseUrl: process.env.LLAMA_CLOUD_BASE_URL,
     apiKey: process.env.LLAMA_CLOUD_API_KEY,
   });
@@ -19,10 +19,10 @@ async function main() {
   while (true) {
     const query = await rl.question("User: ");
     const stream = await chatEngine.chat({ message: query, stream: true });
-    console.log();
     for await (const chunk of stream) {
       process.stdout.write(chunk.response);
     }
+    process.stdout.write("\n");
   }
 }
 

examples/milvus/README.md

@@ -25,10 +25,10 @@ Here are two sample scripts which work with loading and querying data from a Mil
 
 This sample loads the same dataset of movie reviews as sample dataset. You can install https://github.com/zilliztech/attu to inspect the loaded data.
 
-run `npx ts-node milvus/load`
+run `npx tsx milvus/load`
 
 ## Use RAG to Query the data
 
 Check out your data in Attu and change the sample query as you see fit.
 
-run `npx ts-node milvus/query`
+run `npx tsx milvus/query`

examples/mongodb/README.md

@@ -34,7 +34,7 @@ MONGODB_COLLECTION=tiny_tweets_collection
 You are now ready to import our ready-made data set into Mongo. This is the file `tinytweets.json`, a selection of approximately 1000 tweets from @seldo on Twitter in mid-2019. With your environment set up you can do this by running
 
 ```
-npx ts-node mongodb/1_import.ts
+npx tsx mongodb/1_import.ts
 ```
 
 If you don't want to use tweets, you can replace `json_file` with any other array of JSON objects, but you will need to modify some code later to make sure the correct field gets indexed. There is no LlamaIndex-specific code here; you can load your data into Mongo any way you want to.
@@ -59,7 +59,7 @@ MONGODB_VECTOR_INDEX=tiny_tweets_vector_index
 If the data you're indexing is the tweets we gave you, you're ready to go:
 
 ```bash
-npx ts-node mongodb/2_load_and_index.ts
+npx tsx mongodb/2_load_and_index.ts
 ```
 
 > Note: this script is running a couple of minutes and currently doesn't show any progress.
@@ -112,7 +112,7 @@ Now you're ready to query your data!
 You can do this by running
 
 ```bash
-npx ts-node mongodb/3_query.ts
+npx tsx mongodb/3_query.ts
 ```
 
 This sets up a connection to Atlas just like `2_load_and_index.ts` did, then it creates a [query engine](https://docs.llamaindex.ai/en/stable/understanding/querying/querying.html#getting-started) and runs a query against it.

examples/multimodal/context.ts

@@ -4,7 +4,6 @@ import {
   NodeWithScore,
   ObjectType,
   OpenAI,
-  RetrievalEndEvent,
   Settings,
   VectorStoreIndex,
 } from "llamaindex";
@@ -18,8 +17,8 @@ Settings.chunkOverlap = 20;
 Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 
 // Update callbackManager
-Settings.callbackManager.on("retrieve-end", (event: RetrievalEndEvent) => {
-  const { nodes, query } = event.detail.payload;
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes, query } = event.detail;
   const imageNodes = nodes.filter(
     (node: NodeWithScore) => node.node.type === ObjectType.IMAGE_DOCUMENT,
   );

examples/multimodal/rag.ts

@@ -1,7 +1,6 @@
 import {
   MultiModalResponseSynthesizer,
   OpenAI,
-  RetrievalEndEvent,
   Settings,
   VectorStoreIndex,
 } from "llamaindex";
@@ -15,8 +14,8 @@ Settings.chunkOverlap = 20;
 Settings.llm = new OpenAI({ model: "gpt-4-turbo", maxTokens: 512 });
 
 // Update callbackManager
-Settings.callbackManager.on("retrieve-end", (event: RetrievalEndEvent) => {
-  const { nodes, query } = event.detail.payload;
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes, query } = event.detail;
   console.log(`Retrieved ${nodes.length} nodes for query: ${query}`);
 });
 

examples/package.json

@@ -1,10 +1,12 @@
 {
   "name": "@llamaindex/examples",
   "private": true,
-  "version": "0.0.6",
+  "version": "0.0.7",
   "dependencies": {
     "@aws-crypto/sha256-js": "^5.2.0",
+    "@azure/identity": "^4.2.1",
     "@datastax/astra-db-ts": "^1.2.1",
+    "@llamaindex/core": "^0.1.0",
     "@notionhq/client": "^2.2.15",
     "@pinecone-database/pinecone": "^2.2.2",
     "@zilliz/milvus2-sdk-node": "^2.4.2",
@@ -12,13 +14,12 @@
     "commander": "^12.1.0",
     "dotenv": "^16.4.5",
     "js-tiktoken": "^1.0.12",
-    "llamaindex": "^0.4.3",
+    "llamaindex": "^0.5.0",
     "mongodb": "^6.7.0",
     "pathe": "^1.1.2"
   },
   "devDependencies": {
     "@types/node": "^20.14.1",
-    "ts-node": "^10.9.2",
     "tsx": "^4.15.6",
     "typescript": "^5.5.2"
   },

examples/pg-vector-store/README.md

@@ -37,7 +37,7 @@ Read and follow the instructions in the README.md file located one directory up
 
 To import documents and save the embedding vectors to your database:
 
-> `npx ts-node pg-vector-store/load-docs.ts data`
+> `npx tsx pg-vector-store/load-docs.ts data`
 
 where data is the directory containing your input files. Using the `data` directory in the example above will read all of the files in that directory using the LlamaIndexTS default readers for each file type.
 
@@ -45,6 +45,6 @@ where data is the directory containing your input files. Using the `data` direct
 
 To query using the resulting vector store:
 
-> `npx ts-node pg-vector-store/query.ts`
+> `npx tsx pg-vector-store/query.ts`
 
 The script will prompt for a question, then process and present the answer using the PGVectorStore data and your OpenAI API key. It will continue to prompt until you enter `q`, `quit` or `exit` as the next query.

examples/pinecone-vector-store/README.md

@@ -19,7 +19,7 @@ Read and follow the instructions in the README.md file located one directory up
 
 To import documents and save the embedding vectors to your database:
 
-> `npx ts-node pinecone-vector-store/load-docs.ts data`
+> `npx tsx pinecone-vector-store/load-docs.ts data`
 
 where data is the directory containing your input files. Using the _data_ directory in the example above will read all of the files in that directory using the llamaindexTS default readers for each file type.
 
@@ -29,6 +29,6 @@ where data is the directory containing your input files. Using the _data_ direct
 
 To query using the resulting vector store:
 
-> `npx ts-node pinecone-vector-store/query.ts`
+> `npx tsx pinecone-vector-store/query.ts`
 
 The script will prompt for a question, then process and present the answer using the PineconeVectorStore data and your OpenAI API key. It will continue to prompt until you enter `q`, `quit` or `exit` as the next query.

examples/qdrantdb/README.md

@@ -8,4 +8,4 @@ Add your OpenAI API Key into a file called `.env` in the parent folder of this d
 OPEN_API_KEY=sk-you-key
 ```
 
-Now, open a new terminal window and inside `examples`, run `npx ts-node qdrantdb/preFilters.ts`.
+Now, open a new terminal window and inside `examples`, run `npx tsx qdrantdb/preFilters.ts`.

examples/qdrantdb/preFilters.ts

@@ -1,8 +1,8 @@
 import * as dotenv from "dotenv";
 import {
-  CallbackManager,
   Document,
   MetadataMode,
+  NodeWithScore,
   QdrantVectorStore,
   Settings,
   VectorStoreIndex,
@@ -10,13 +10,12 @@ import {
 } from "llamaindex";
 
 // Update callback manager
-Settings.callbackManager = new CallbackManager({
-  onRetrieve: (data) => {
-    console.log(
-      "The retrieved nodes are:",
-      data.nodes.map((node) => node.node.getContent(MetadataMode.NONE)),
-    );
-  },
+Settings.callbackManager.on("retrieve-end", (event) => {
+  const { nodes } = event.detail;
+  console.log(
+    "The retrieved nodes are:",
+    nodes.map((node: NodeWithScore) => node.node.getContent(MetadataMode.NONE)),
+  );
 });
 
 // Load environment variables from local .env file

examples/recipes/cost-analysis.ts

@@ -1,7 +1,7 @@
+import { extractText } from "@llamaindex/core/utils";
 import { encodingForModel } from "js-tiktoken";
-import { ChatMessage, OpenAI, type LLMStartEvent } from "llamaindex";
+import { ChatMessage, OpenAI } from "llamaindex";
 import { Settings } from "llamaindex/Settings";
-import { extractText } from "llamaindex/llm/utils";
 
 const encoding = encodingForModel("gpt-4-0125-preview");
 
@@ -12,8 +12,8 @@ const llm = new OpenAI({
 
 let tokenCount = 0;
 
-Settings.callbackManager.on("llm-start", (event: LLMStartEvent) => {
-  const { messages } = event.detail.payload;
+Settings.callbackManager.on("llm-start", (event) => {
+  const { messages } = event.detail;
   messages.reduce((count: number, message: ChatMessage) => {
     return count + encoding.encode(extractText(message.content)).length;
   }, 0);
@@ -24,7 +24,7 @@ Settings.callbackManager.on("llm-start", (event: LLMStartEvent) => {
 });
 
 Settings.callbackManager.on("llm-stream", (event) => {
-  const { chunk } = event.detail.payload;
+  const { chunk } = event.detail;
   const { delta } = chunk;
   tokenCount += encoding.encode(extractText(delta)).length;
   if (tokenCount > 20) {

package.json

@@ -39,9 +39,6 @@
       "trim": "1.0.1",
       "@babel/traverse": "7.23.2",
       "protobufjs": "7.2.6"
-    },
-    "patchedDependencies": {
-      "bunchee@5.2.1": "patches/bunchee@5.2.1.patch"
     }
   },
   "lint-staged": {

packages/autotool/CHANGELOG.md

@@ -1,5 +1,14 @@
 # @llamaindex/autotool
 
+## 2.0.0
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
 ## 1.0.0
 
 ### Patch Changes

packages/autotool/examples/02_nextjs/CHANGELOG.md

@@ -1,5 +1,81 @@
 # @llamaindex/autotool-02-next-example
 
+## 0.1.25
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+  - @llamaindex/autotool@2.0.0
+
+## 0.1.24
+
+### Patch Changes
+
+- llamaindex@0.4.14
+- @llamaindex/autotool@1.0.0
+
+## 0.1.23
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.22
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.21
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.20
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.19
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.18
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+  - @llamaindex/autotool@1.0.0
+
+## 0.1.17
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+  - @llamaindex/autotool@1.0.0
+
 ## 0.1.16
 
 ### Patch Changes

packages/autotool/examples/02_nextjs/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool-02-next-example",
   "private": true,
-  "version": "0.1.16",
+  "version": "0.1.25",
   "scripts": {
     "dev": "next dev",
     "build": "next build",

packages/autotool/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/autotool",
   "type": "module",
-  "version": "1.0.0",
+  "version": "2.0.0",
   "description": "auto transpile your JS function to LLM Agent compatible",
   "files": [
     "dist",
@@ -51,7 +51,7 @@
     "unplugin": "^1.10.1"
   },
   "peerDependencies": {
-    "llamaindex": "^0.4.6",
+    "llamaindex": "^0.5.0",
     "openai": "^4",
     "typescript": "^4"
   },
@@ -70,7 +70,7 @@
     "@swc/types": "^0.1.8",
     "@types/json-schema": "^7.0.15",
     "@types/node": "^20.12.11",
-    "bunchee": "^5.2.1",
+    "bunchee": "5.3.0-beta.0",
     "llamaindex": "workspace:*",
     "next": "14.2.3",
     "rollup": "^4.18.0",

packages/cloud/CHANGELOG.md

@@ -0,0 +1,25 @@
+# @llamaindex/cloud
+
+## 0.1.4
+
+### Patch Changes
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+
+## 0.1.3
+
+### Patch Changes
+
+- 1c444d5: feat(cloud): update openapi.json
+
+## 0.1.2
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
+## 0.1.1
+
+### Patch Changes
+
+- 321c39d: fix: generate api as class

packages/cloud/README.md

@@ -5,7 +5,7 @@
 ## Usage
 
 ```ts
-import { OpenAPI, Service } from "@llamaindex/cloud/api";
+import { OpenAPI } from "@llamaindex/cloud/api";
 OpenAPI.TOKEN = "YOUR_API_KEY";
 OpenAPI.BASE = "https://api.cloud.llamaindex.ai/";
 // ...

packages/cloud/openapi-ts.config.ts

@@ -9,6 +9,9 @@ export default defineConfig({
     format: "prettier",
     lint: "eslint",
   },
+  services: {
+    asClass: true,
+  },
   types: {
     enums: "javascript",
   },

packages/cloud/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloud",
-  "version": "0.1.0",
+  "version": "0.1.4",
   "type": "module",
   "license": "MIT",
   "scripts": {
@@ -35,6 +35,6 @@
   },
   "devDependencies": {
     "@hey-api/openapi-ts": "^0.48.0",
-    "bunchee": "^5.2.1"
+    "bunchee": "5.3.0-beta.0"
   }
 }

packages/cloud/src/api.ts

@@ -1,4 +1 @@
-import * as Service from "./client/services.gen";
-
 export * from "./client";
-export { Service };

packages/community/CHANGELOG.md

@@ -1,5 +1,73 @@
 # @llamaindex/community
 
+## 0.0.19
+
+### Patch Changes
+
+- 16ef5dd: refactor: depends on core pacakge instead of llamaindex
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+  - @llamaindex/core@0.1.0
+
+## 0.0.18
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.17
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.16
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.14
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.10
 
 ### Patch Changes

packages/community/README.md

@@ -5,6 +5,7 @@
 ## Current Features:
 
 - Bedrock support for the Anthropic Claude Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
+- Bedrock support for the Meta LLama 2 and 3 Models [usage](https://ts.llamaindex.ai/modules/llms/available_llms/bedrock)
 
 ## LICENSE
 

packages/community/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/community",
   "description": "Community package for LlamaIndexTS",
-  "version": "0.0.10",
+  "version": "0.0.19",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",
@@ -38,20 +38,15 @@
     "directory": "packages/community"
   },
   "scripts": {
-    "build": "rm -rf ./dist && pnpm run build:code && pnpm run build:type",
-    "build:code": "tsup",
-    "build:type": "tsc -p tsconfig.json",
-    "dev": "concurrently \"pnpm run build:esm --watch\" \"pnpm run build:cjs --watch\" \"pnpm run build:type --watch\""
+    "build": "bunchee",
+    "dev": "bunchee --watch"
   },
   "devDependencies": {
-    "@swc/cli": "^0.3.12",
-    "@swc/core": "^1.6.3",
-    "concurrently": "^8.2.2",
-    "tsup": "^8.1.0"
+    "@types/node": "^20.14.2",
+    "bunchee": "5.3.0-beta.0"
   },
   "dependencies": {
     "@aws-sdk/client-bedrock-runtime": "^3.600.0",
-    "@types/node": "^20.14.2",
-    "llamaindex": "workspace:*"
+    "@llamaindex/core": "workspace:*"
   }
 }

packages/community/src/llm/bedrock/base.ts

@@ -1,53 +1,35 @@
 import {
   BedrockRuntimeClient,
+  type BedrockRuntimeClientConfig,
   InvokeModelCommand,
   InvokeModelWithResponseStreamCommand,
-  ResponseStream,
-  type BedrockRuntimeClientConfig,
-  type InvokeModelCommandInput,
-  type InvokeModelWithResponseStreamCommandInput,
 } from "@aws-sdk/client-bedrock-runtime";
-import type {
-  BaseTool,
-  ChatMessage,
-  ChatResponse,
-  ChatResponseChunk,
-  CompletionResponse,
-  LLMChatParamsNonStreaming,
-  LLMChatParamsStreaming,
-  LLMCompletionParamsNonStreaming,
-  LLMCompletionParamsStreaming,
-  LLMMetadata,
-  PartialToolCall,
-  ToolCall,
-  ToolCallLLMMessageOptions,
-} from "llamaindex";
-import { ToolCallLLM, streamConverter, wrapLLMEvent } from "llamaindex";
-import type {
-  AnthropicNoneStreamingResponse,
-  AnthropicTextContent,
-  StreamEvent,
-  ToolBlock,
-  ToolChoice,
-} from "./types.js";
 import {
-  mapBaseToolsToAnthropicTools,
-  mapChatMessagesToAnthropicMessages,
-  mapMessageContentToMessageContentDetails,
-  toUtf8,
-} from "./utils.js";
-
-export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+  type ChatMessage,
+  type ChatResponse,
+  type CompletionResponse,
+  type LLMChatParamsNonStreaming,
+  type LLMChatParamsStreaming,
+  type LLMCompletionParamsNonStreaming,
+  type LLMCompletionParamsStreaming,
+  type LLMMetadata,
+  ToolCallLLM,
+  type ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import { streamConverter, wrapLLMEvent } from "@llamaindex/core/utils";
+import {
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "./provider";
+import { PROVIDERS } from "./providers";
+import { mapMessageContentToMessageContentDetails } from "./utils.js";
 
 export type BedrockChatParamsStreaming = LLMChatParamsStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
 >;
 
-export type BedrockChatStreamResponse = AsyncIterable<
-  ChatResponseChunk<ToolCallLLMMessageOptions>
->;
-
 export type BedrockChatParamsNonStreaming = LLMChatParamsNonStreaming<
   BedrockAdditionalChatOptions,
   ToolCallLLMMessageOptions
@@ -151,174 +133,6 @@ export const TOOL_CALL_MODELS = [
   BEDROCK_MODELS.ANTHROPIC_CLAUDE_3_5_SONNET,
 ];
 
-abstract class Provider<ProviderStreamEvent extends {} = {}> {
-  abstract getTextFromResponse(response: Record<string, any>): string;
-
-  abstract getToolsFromResponse<T extends {} = {}>(
-    response: Record<string, any>,
-  ): T[];
-
-  getStreamingEventResponse(
-    response: Record<string, any>,
-  ): ProviderStreamEvent | undefined {
-    return response.chunk?.bytes
-      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
-      : undefined;
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    yield* streamConverter(stream, (response) => {
-      return {
-        delta: this.getTextFromStreamResponse(response),
-        raw: response,
-      };
-    });
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    return this.getTextFromResponse(response);
-  }
-
-  abstract getRequestBody<T extends ChatMessage>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
-}
-
-class AnthropicProvider extends Provider<StreamEvent> {
-  getResultFromResponse(
-    response: Record<string, any>,
-  ): AnthropicNoneStreamingResponse {
-    return JSON.parse(toUtf8(response.body));
-  }
-
-  getToolsFromResponse<AnthropicToolContent>(
-    response: Record<string, any>,
-  ): AnthropicToolContent[] {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "tool_use")
-      .map((item) => item as AnthropicToolContent);
-  }
-
-  getTextFromResponse(response: Record<string, any>): string {
-    const result = this.getResultFromResponse(response);
-    return result.content
-      .filter((item) => item.type === "text")
-      .map((item) => (item as AnthropicTextContent).text)
-      .join(" ");
-  }
-
-  getTextFromStreamResponse(response: Record<string, any>): string {
-    const event = this.getStreamingEventResponse(response);
-    if (event?.type === "content_block_delta") {
-      if (event.delta.type === "text_delta") return event.delta.text;
-      if (event.delta.type === "input_json_delta")
-        return event.delta.partial_json;
-    }
-    return "";
-  }
-
-  async *reduceStream(
-    stream: AsyncIterable<ResponseStream>,
-  ): BedrockChatStreamResponse {
-    let collecting = [];
-    let tool: ToolBlock | undefined = undefined;
-    // #TODO this should be broken down into a separate consumer
-    for await (const response of stream) {
-      const event = this.getStreamingEventResponse(response);
-      if (
-        event?.type === "content_block_start" &&
-        event.content_block.type === "tool_use"
-      ) {
-        tool = event.content_block;
-        continue;
-      }
-
-      if (
-        event?.type === "content_block_delta" &&
-        event.delta.type === "input_json_delta"
-      ) {
-        collecting.push(event.delta.partial_json);
-      }
-
-      let options: undefined | ToolCallLLMMessageOptions = undefined;
-      if (tool && collecting.length) {
-        const input = collecting.filter((item) => item).join("");
-        // We have all we need to parse the tool_use json
-        if (event?.type === "content_block_stop") {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input: JSON.parse(input),
-              } as ToolCall,
-            ],
-          };
-          // reset the collection/tool
-          collecting = [];
-          tool = undefined;
-        } else {
-          options = {
-            toolCall: [
-              {
-                id: tool.id,
-                name: tool.name,
-                input,
-              } as PartialToolCall,
-            ],
-          };
-        }
-      }
-      const delta = this.getTextFromStreamResponse(response);
-      if (!delta && !options) continue;
-
-      yield {
-        delta,
-        options,
-        raw: response,
-      };
-    }
-  }
-
-  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
-    metadata: LLMMetadata,
-    messages: T[],
-    tools?: BaseTool[],
-    options?: BedrockAdditionalChatOptions,
-  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
-    const extra: Record<string, unknown> = {};
-    if (options?.toolChoice) {
-      extra["tool_choice"] = options?.toolChoice;
-    }
-    const mapped = mapChatMessagesToAnthropicMessages(messages);
-    return {
-      modelId: metadata.model,
-      contentType: "application/json",
-      accept: "application/json",
-      body: JSON.stringify({
-        anthropic_version: "bedrock-2023-05-31",
-        messages: mapped,
-        tools: mapBaseToolsToAnthropicTools(tools),
-        max_tokens: metadata.maxTokens,
-        temperature: metadata.temperature,
-        top_p: metadata.topP,
-        ...extra,
-      }),
-    };
-  }
-}
-
-// Other providers could go here
-const PROVIDERS: { [key: string]: Provider } = {
-  anthropic: new AnthropicProvider(),
-};
-
 const getProvider = (model: string): Provider => {
   const providerName = model.split(".")[0];
   if (!(providerName in PROVIDERS)) {
@@ -373,6 +187,10 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     this.temperature = temperature ?? DEFAULT_BEDROCK_PARAMS.temperature;
     this.topP = topP ?? DEFAULT_BEDROCK_PARAMS.topP;
     this.client = new BedrockRuntimeClient(params);
+
+    if (!this.supportToolCall) {
+      console.warn(`The model "${this.model}" doesn't support ToolCall`);
+    }
   }
 
   get supportToolCall(): boolean {
@@ -402,10 +220,13 @@ export class Bedrock extends ToolCallLLM<BedrockAdditionalChatOptions> {
     );
     const command = new InvokeModelCommand(input);
     const response = await this.client.send(command);
-    const tools = this.provider.getToolsFromResponse(response);
-    const options: ToolCallLLMMessageOptions = tools.length
-      ? { toolCall: tools }
-      : {};
+    let options: ToolCallLLMMessageOptions = {};
+    if (this.supportToolCall) {
+      const tools = this.provider.getToolsFromResponse(response);
+      if (tools.length) {
+        options = { toolCall: tools };
+      }
+    }
     return {
       raw: response,
       message: {

packages/community/src/llm/bedrock/provider.ts

@@ -0,0 +1,59 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import {
+  type BaseTool,
+  type ChatMessage,
+  type ChatResponseChunk,
+  type LLMMetadata,
+  type ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import { streamConverter } from "@llamaindex/core/utils";
+import type { ToolChoice } from "./types";
+import { toUtf8 } from "./utils";
+
+export type BedrockAdditionalChatOptions = { toolChoice: ToolChoice };
+
+export type BedrockChatStreamResponse = AsyncIterable<
+  ChatResponseChunk<ToolCallLLMMessageOptions>
+>;
+
+export abstract class Provider<ProviderStreamEvent extends {} = {}> {
+  abstract getTextFromResponse(response: Record<string, any>): string;
+
+  abstract getToolsFromResponse<T extends {} = {}>(
+    response: Record<string, any>,
+  ): T[];
+
+  getStreamingEventResponse(
+    response: Record<string, any>,
+  ): ProviderStreamEvent | undefined {
+    return response.chunk?.bytes
+      ? (JSON.parse(toUtf8(response.chunk?.bytes)) as ProviderStreamEvent)
+      : undefined;
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    yield* streamConverter(stream, (response) => {
+      return {
+        delta: this.getTextFromStreamResponse(response),
+        raw: response,
+      };
+    });
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    return this.getTextFromResponse(response);
+  }
+
+  abstract getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput;
+}

packages/community/src/llm/bedrock/providers/anthropic.ts

@@ -0,0 +1,154 @@
+import {
+  type InvokeModelCommandInput,
+  type InvokeModelWithResponseStreamCommandInput,
+  ResponseStream,
+} from "@aws-sdk/client-bedrock-runtime";
+import type {
+  BaseTool,
+  ChatMessage,
+  LLMMetadata,
+  PartialToolCall,
+  ToolCall,
+  ToolCallLLMMessageOptions,
+} from "@llamaindex/core/llms";
+import {
+  type BedrockAdditionalChatOptions,
+  type BedrockChatStreamResponse,
+  Provider,
+} from "../provider";
+import type {
+  AnthropicNoneStreamingResponse,
+  AnthropicStreamEvent,
+  AnthropicTextContent,
+  ToolBlock,
+} from "../types";
+import {
+  mapBaseToolsToAnthropicTools,
+  mapChatMessagesToAnthropicMessages,
+  toUtf8,
+} from "../utils";
+
+export class AnthropicProvider extends Provider<AnthropicStreamEvent> {
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): AnthropicNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse<AnthropicToolContent>(
+    response: Record<string, any>,
+  ): AnthropicToolContent[] {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "tool_use")
+      .map((item) => item as AnthropicToolContent);
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.content
+      .filter((item) => item.type === "text")
+      .map((item) => (item as AnthropicTextContent).text)
+      .join(" ");
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.type === "content_block_delta") {
+      if (event.delta.type === "text_delta") return event.delta.text;
+      if (event.delta.type === "input_json_delta")
+        return event.delta.partial_json;
+    }
+    return "";
+  }
+
+  async *reduceStream(
+    stream: AsyncIterable<ResponseStream>,
+  ): BedrockChatStreamResponse {
+    let collecting = [];
+    let tool: ToolBlock | undefined = undefined;
+    // #TODO this should be broken down into a separate consumer
+    for await (const response of stream) {
+      const event = this.getStreamingEventResponse(response);
+      if (
+        event?.type === "content_block_start" &&
+        event.content_block.type === "tool_use"
+      ) {
+        tool = event.content_block;
+        continue;
+      }
+
+      if (
+        event?.type === "content_block_delta" &&
+        event.delta.type === "input_json_delta"
+      ) {
+        collecting.push(event.delta.partial_json);
+      }
+
+      let options: undefined | ToolCallLLMMessageOptions = undefined;
+      if (tool && collecting.length) {
+        const input = collecting.filter((item) => item).join("");
+        // We have all we need to parse the tool_use json
+        if (event?.type === "content_block_stop") {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input: JSON.parse(input),
+              } as ToolCall,
+            ],
+          };
+          // reset the collection/tool
+          collecting = [];
+          tool = undefined;
+        } else {
+          options = {
+            toolCall: [
+              {
+                id: tool.id,
+                name: tool.name,
+                input,
+              } as PartialToolCall,
+            ],
+          };
+        }
+      }
+      const delta = this.getTextFromStreamResponse(response);
+      if (!delta && !options) continue;
+
+      yield {
+        delta,
+        options,
+        raw: response,
+      };
+    }
+  }
+
+  getRequestBody<T extends ChatMessage<ToolCallLLMMessageOptions>>(
+    metadata: LLMMetadata,
+    messages: T[],
+    tools?: BaseTool[],
+    options?: BedrockAdditionalChatOptions,
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    const extra: Record<string, unknown> = {};
+    if (options?.toolChoice) {
+      extra["tool_choice"] = options?.toolChoice;
+    }
+    const mapped = mapChatMessagesToAnthropicMessages(messages);
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        anthropic_version: "bedrock-2023-05-31",
+        messages: mapped,
+        tools: mapBaseToolsToAnthropicTools(tools),
+        max_tokens: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+        ...extra,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/providers/index.ts

@@ -0,0 +1,9 @@
+import { Provider } from "../provider";
+import { AnthropicProvider } from "./anthropic";
+import { MetaProvider } from "./meta";
+
+// Other providers should go here
+export const PROVIDERS: { [key: string]: Provider } = {
+  anthropic: new AnthropicProvider(),
+  meta: new MetaProvider(),
+};

packages/community/src/llm/bedrock/providers/meta.ts

@@ -0,0 +1,69 @@
+import type {
+  InvokeModelCommandInput,
+  InvokeModelWithResponseStreamCommandInput,
+} from "@aws-sdk/client-bedrock-runtime";
+import type { ChatMessage, LLMMetadata } from "@llamaindex/core/llms";
+import type { MetaNoneStreamingResponse, MetaStreamEvent } from "../types";
+import {
+  mapChatMessagesToMetaLlama2Messages,
+  mapChatMessagesToMetaLlama3Messages,
+  toUtf8,
+} from "../utils";
+
+import { Provider } from "../provider";
+
+export class MetaProvider extends Provider<MetaStreamEvent> {
+  constructor() {
+    super();
+  }
+
+  getResultFromResponse(
+    response: Record<string, any>,
+  ): MetaNoneStreamingResponse {
+    return JSON.parse(toUtf8(response.body));
+  }
+
+  getToolsFromResponse(_response: Record<string, any>): never {
+    throw new Error("Not supported by this provider.");
+  }
+
+  getTextFromResponse(response: Record<string, any>): string {
+    const result = this.getResultFromResponse(response);
+    return result.generation;
+  }
+
+  getTextFromStreamResponse(response: Record<string, any>): string {
+    const event = this.getStreamingEventResponse(response);
+    if (event?.generation) {
+      return event.generation;
+    }
+    return "";
+  }
+
+  getRequestBody<T extends ChatMessage>(
+    metadata: LLMMetadata,
+    messages: T[],
+  ): InvokeModelCommandInput | InvokeModelWithResponseStreamCommandInput {
+    let promptFunction: (messages: ChatMessage[]) => string;
+
+    if (metadata.model.startsWith("meta.llama3")) {
+      promptFunction = mapChatMessagesToMetaLlama3Messages;
+    } else if (metadata.model.startsWith("meta.llama2")) {
+      promptFunction = mapChatMessagesToMetaLlama2Messages;
+    } else {
+      throw new Error(`Meta model ${metadata.model} is not supported`);
+    }
+
+    return {
+      modelId: metadata.model,
+      contentType: "application/json",
+      accept: "application/json",
+      body: JSON.stringify({
+        prompt: promptFunction(messages),
+        max_gen_len: metadata.maxTokens,
+        temperature: metadata.temperature,
+        top_p: metadata.topP,
+      }),
+    };
+  }
+}

packages/community/src/llm/bedrock/types.ts

@@ -79,7 +79,7 @@ export type ToolChoice =
   | { type: "auto" }
   | { type: "tool"; name: string };
 
-export type StreamEvent =
+export type AnthropicStreamEvent =
   | { type: "message_start"; message: Message }
   | ContentBlockStart
   | ContentBlockDelta
@@ -93,6 +93,8 @@ export type AnthropicContent =
   | AnthropicToolContent
   | AnthropicToolResultContent;
 
+export type MetaTextContent = string;
+
 export type AnthropicTextContent = {
   type: "text";
   text: string;
@@ -133,6 +135,11 @@ export type AnthropicMessage = {
   content: AnthropicContent[];
 };
 
+export type MetaMessage = {
+  role: "user" | "assistant" | "system";
+  content: MetaTextContent;
+};
+
 export type AnthropicNoneStreamingResponse = {
   id: string;
   type: "message";
@@ -143,3 +150,16 @@ export type AnthropicNoneStreamingResponse = {
   stop_sequence?: string;
   usage: { input_tokens: number; output_tokens: number };
 };
+
+type MetaResponse = {
+  generation: string;
+  prompt_token_count: number;
+  generation_token_count: number;
+  stop_reason: "stop" | "length";
+};
+
+export type MetaStreamEvent = MetaResponse & {
+  "amazon-bedrock-invocationMetrics": InvocationMetrics;
+};
+
+export type MetaNoneStreamingResponse = MetaResponse;

packages/community/src/llm/bedrock/utils.ts

@@ -1,18 +1,20 @@
+import type { JSONObject } from "@llamaindex/core/global";
 import type {
   BaseTool,
   ChatMessage,
-  JSONObject,
   MessageContent,
   MessageContentDetail,
+  MessageContentTextDetail,
   ToolCallLLMMessageOptions,
   ToolMetadata,
-} from "llamaindex";
+} from "@llamaindex/core/llms";
 import type {
   AnthropicContent,
   AnthropicImageContent,
   AnthropicMediaTypes,
   AnthropicMessage,
   AnthropicTextContent,
+  MetaMessage,
 } from "./types.js";
 
 const ACCEPTED_IMAGE_MIME_TYPES = [
@@ -148,6 +150,85 @@ export const mapChatMessagesToAnthropicMessages = <
   return mergeNeighboringSameRoleMessages(mapped);
 };
 
+export const mapChatMessagesToMetaMessages = <T extends ChatMessage>(
+  messages: T[],
+): MetaMessage[] => {
+  return messages.map((msg) => {
+    let content: string;
+    if (typeof msg.content === "string") {
+      content = msg.content;
+    } else {
+      content = (msg.content[0] as MessageContentTextDetail).text;
+    }
+    return {
+      role:
+        msg.role === "assistant"
+          ? "assistant"
+          : msg.role === "user"
+            ? "user"
+            : "system",
+      content,
+    };
+  });
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-3
+ */
+export const mapChatMessagesToMetaLlama3Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages).map((message) => {
+    const text = message.content;
+    return `<|start_header_id|>${message.role}<|end_header_id|>\n${text}\n<|eot_id|>\n`;
+  });
+  return (
+    "<|begin_of_text|>" +
+    mapped.join("\n") +
+    "\n<|start_header_id|>assistant<|end_header_id|>\n"
+  );
+};
+
+/**
+ * Documentation at https://llama.meta.com/docs/model-cards-and-prompt-formats/meta-llama-2
+ */
+export const mapChatMessagesToMetaLlama2Messages = <T extends ChatMessage>(
+  messages: T[],
+): string => {
+  const mapped = mapChatMessagesToMetaMessages(messages);
+  let output = "<s>";
+  let insideInst = false;
+  let needsStartAgain = false;
+  for (const message of mapped) {
+    if (needsStartAgain) {
+      output += "<s>";
+      needsStartAgain = false;
+    }
+    const text = message.content;
+    if (message.role === "system") {
+      if (!insideInst) {
+        output += "[INST] ";
+        insideInst = true;
+      }
+      output += `<<SYS>>\n${text}\n<</SYS>>\n`;
+    } else if (message.role === "user") {
+      output += text;
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+    } else if (message.role === "assistant") {
+      if (insideInst) {
+        output += " [/INST]";
+        insideInst = false;
+      }
+      output += ` ${text} </s>\n`;
+      needsStartAgain = true;
+    }
+  }
+  return output;
+};
+
 export const mapTextContent = (text: string): AnthropicTextContent => {
   return { type: "text", text };
 };

packages/community/tsconfig.script.json

@@ -1,9 +0,0 @@
-{
-  "extends": "../../tsconfig.json",
-  "compilerOptions": {
-    "outDir": "./dist/script/type",
-    "tsBuildInfoFile": "./dist/script/.tsbuildinfo",
-    "emitDeclarationOnly": true
-  },
-  "include": ["./tsup.config.ts"]
-}

packages/community/tsup.config.ts

@@ -1,9 +0,0 @@
-import { defineConfig } from "tsup";
-
-export default defineConfig([
-  {
-    entry: ["src/index.ts", "src/llm/bedrock/base.ts"],
-    format: ["cjs", "esm"],
-    sourcemap: true,
-  },
-]);

packages/core/CHANGELOG.md

@@ -0,0 +1,32 @@
+# @llamaindex/core
+
+## 0.1.0
+
+### Minor Changes
+
+- 16ef5dd: refactor: simplify callback manager
+
+  Change `event.detail.payload` to `event.detail`
+
+### Patch Changes
+
+- 16ef5dd: refactor: move callback manager & llm to core module
+
+  For people who import `llamaindex/llms/base` or `llamaindex/llms/utils`,
+  use `@llamaindex/core/llms` and `@llamaindex/core/utils` instead.
+
+## 0.0.3
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/env@0.1.8
+
+## 0.0.2
+
+### Patch Changes
+
+- f10b41d: fix: release files
+- Updated dependencies [41fe871]
+  - @llamaindex/env@0.1.7

packages/core/package.json

@@ -1,9 +1,23 @@
 {
   "name": "@llamaindex/core",
   "type": "module",
-  "version": "0.0.1",
+  "version": "0.1.0",
   "description": "LlamaIndex Core Module",
   "exports": {
+    "./llms": {
+      "require": {
+        "types": "./dist/llms/index.d.cts",
+        "default": "./dist/llms/index.cjs"
+      },
+      "import": {
+        "types": "./dist/llms/index.d.ts",
+        "default": "./dist/llms/index.js"
+      },
+      "default": {
+        "types": "./dist/llms/index.d.ts",
+        "default": "./dist/llms/index.js"
+      }
+    },
     "./decorator": {
       "require": {
         "types": "./dist/decorator/index.d.cts",
@@ -45,8 +59,25 @@
         "types": "./dist/schema/index.d.ts",
         "default": "./dist/schema/index.js"
       }
+    },
+    "./utils": {
+      "require": {
+        "types": "./dist/utils/index.d.cts",
+        "default": "./dist/utils/index.cjs"
+      },
+      "import": {
+        "types": "./dist/utils/index.d.ts",
+        "default": "./dist/utils/index.js"
+      },
+      "default": {
+        "types": "./dist/utils/index.d.ts",
+        "default": "./dist/utils/index.js"
+      }
     }
   },
+  "files": [
+    "dist"
+  ],
   "scripts": {
     "dev": "bunchee --watch",
     "build": "bunchee"
@@ -57,7 +88,8 @@
     "url": "https://github.com/himself65/LlamaIndexTS.git"
   },
   "devDependencies": {
-    "bunchee": "^5.2.1"
+    "ajv": "^8.16.0",
+    "bunchee": "5.3.0-beta.0"
   },
   "dependencies": {
     "@llamaindex/env": "workspace:*",

packages/core/src/global/index.ts

@@ -1 +1,11 @@
 export { Settings } from "./settings";
+export { CallbackManager } from "./settings/callback-manager";
+export type {
+  LLMEndEvent,
+  LLMStartEvent,
+  LLMStreamEvent,
+  LLMToolCallEvent,
+  LLMToolResultEvent,
+  LlamaIndexEventMaps,
+} from "./settings/callback-manager";
+export type { JSONArray, JSONObject, JSONValue } from "./type";

packages/core/src/global/settings.ts

@@ -1,3 +1,9 @@
+import {
+  type CallbackManager,
+  getCallbackManager,
+  setCallbackManager,
+  withCallbackManager,
+} from "./settings/callback-manager";
 import {
   getChunkSize,
   setChunkSize,
@@ -14,4 +20,19 @@ export const Settings = {
   withChunkSize<Result>(chunkSize: number, fn: () => Result): Result {
     return withChunkSize(chunkSize, fn);
   },
+
+  get callbackManager(): CallbackManager {
+    return getCallbackManager();
+  },
+
+  set callbackManager(callbackManager: CallbackManager) {
+    setCallbackManager(callbackManager);
+  },
+
+  withCallbackManager<Result>(
+    callbackManager: CallbackManager,
+    fn: () => Result,
+  ): Result {
+    return withCallbackManager(callbackManager, fn);
+  },
 };

packages/core/src/global/settings/callback-manager.ts

@@ -0,0 +1,140 @@
+import { AsyncLocalStorage, CustomEvent } from "@llamaindex/env";
+import type {
+  ChatMessage,
+  ChatResponse,
+  ChatResponseChunk,
+  ToolCall,
+  ToolOutput,
+} from "../../llms";
+import { EventCaller, getEventCaller } from "../../utils/event-caller";
+import type { UUID } from "../type";
+
+export type LLMStartEvent = {
+  id: UUID;
+  messages: ChatMessage[];
+};
+
+export type LLMToolCallEvent = {
+  toolCall: ToolCall;
+};
+
+export type LLMToolResultEvent = {
+  toolCall: ToolCall;
+  toolResult: ToolOutput;
+};
+
+export type LLMEndEvent = {
+  id: UUID;
+  response: ChatResponse;
+};
+
+export type LLMStreamEvent = {
+  id: UUID;
+  chunk: ChatResponseChunk;
+};
+
+export interface LlamaIndexEventMaps {
+  "llm-start": LLMStartEvent;
+  "llm-end": LLMEndEvent;
+  "llm-tool-call": LLMToolCallEvent;
+  "llm-tool-result": LLMToolResultEvent;
+  "llm-stream": LLMStreamEvent;
+}
+
+export class LlamaIndexCustomEvent<T = any> extends CustomEvent<T> {
+  reason: EventCaller | null = null;
+  private constructor(
+    event: string,
+    options?: CustomEventInit & {
+      reason?: EventCaller | null;
+    },
+  ) {
+    super(event, options);
+    this.reason = options?.reason ?? null;
+  }
+
+  static fromEvent<Type extends keyof LlamaIndexEventMaps>(
+    type: Type,
+    detail: LlamaIndexEventMaps[Type],
+  ) {
+    return new LlamaIndexCustomEvent(type, {
+      detail: detail,
+      reason: getEventCaller(),
+    });
+  }
+}
+
+type EventHandler<Event> = (event: LlamaIndexCustomEvent<Event>) => void;
+
+export class CallbackManager {
+  #handlers = new Map<keyof LlamaIndexEventMaps, EventHandler<any>[]>();
+
+  on<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    handler: EventHandler<LlamaIndexEventMaps[K]>,
+  ) {
+    if (!this.#handlers.has(event)) {
+      this.#handlers.set(event, []);
+    }
+    this.#handlers.get(event)!.push(handler);
+    return this;
+  }
+
+  off<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    handler: EventHandler<LlamaIndexEventMaps[K]>,
+  ) {
+    if (!this.#handlers.has(event)) {
+      return this;
+    }
+    const cbs = this.#handlers.get(event)!;
+    const index = cbs.indexOf(handler);
+    if (index > -1) {
+      cbs.splice(index, 1);
+    }
+    return this;
+  }
+
+  dispatchEvent<K extends keyof LlamaIndexEventMaps>(
+    event: K,
+    detail: LlamaIndexEventMaps[K],
+  ) {
+    const cbs = this.#handlers.get(event);
+    if (!cbs) {
+      return;
+    }
+    queueMicrotask(() => {
+      cbs.forEach((handler) =>
+        handler(
+          LlamaIndexCustomEvent.fromEvent(event, structuredClone(detail)),
+        ),
+      );
+    });
+  }
+}
+
+export const globalCallbackManager = new CallbackManager();
+
+const callbackManagerAsyncLocalStorage =
+  new AsyncLocalStorage<CallbackManager>();
+
+let currentCallbackManager: CallbackManager | null = null;
+
+export function getCallbackManager(): CallbackManager {
+  return (
+    callbackManagerAsyncLocalStorage.getStore() ??
+    currentCallbackManager ??
+    globalCallbackManager
+  );
+}
+
+export function setCallbackManager(callbackManager: CallbackManager) {
+  currentCallbackManager = callbackManager;
+}
+
+export function withCallbackManager<Result>(
+  callbackManager: CallbackManager,
+  fn: () => Result,
+): Result {
+  return callbackManagerAsyncLocalStorage.run(callbackManager, fn);
+}

packages/core/src/global/type.ts

@@ -0,0 +1,9 @@
+export type UUID = `${string}-${string}-${string}-${string}-${string}`;
+
+export type JSONValue = string | number | boolean | JSONObject | JSONArray;
+
+export type JSONObject = {
+  [key: string]: JSONValue;
+};
+
+export type JSONArray = Array<JSONValue>;

packages/core/src/index.ts

@@ -1 +0,0 @@
-export * from "./schema";

packages/core/src/llms/base.ts

@@ -1,3 +1,5 @@
+import { streamConverter } from "../utils";
+import { extractText } from "../utils/llms";
 import type {
   ChatResponse,
   ChatResponseChunk,
@@ -9,8 +11,7 @@ import type {
   LLMCompletionParamsStreaming,
   LLMMetadata,
   ToolCallLLMMessageOptions,
-} from "./types.js";
-import { extractText, streamConverter } from "./utils.js";
+} from "./type";
 
 export abstract class BaseLLM<
   AdditionalChatOptions extends object = object,

packages/core/src/llms/index.ts

@@ -0,0 +1,32 @@
+export { BaseLLM, ToolCallLLM } from "./base";
+export type {
+  BaseTool,
+  BaseToolWithCall,
+  ChatMessage,
+  ChatResponse,
+  ChatResponseChunk,
+  CompletionResponse,
+  LLM,
+  LLMChat,
+  LLMChatParamsBase,
+  LLMChatParamsNonStreaming,
+  LLMChatParamsStreaming,
+  LLMCompletionParamsBase,
+  LLMCompletionParamsNonStreaming,
+  LLMCompletionParamsStreaming,
+  LLMMetadata,
+  MessageContent,
+  MessageContentDetail,
+  MessageContentImageDetail,
+  MessageContentTextDetail,
+  MessageType,
+  PartialToolCall,
+  TextChatMessage,
+  ToolCall,
+  ToolCallLLMMessageOptions,
+  ToolCallOptions,
+  ToolMetadata,
+  ToolOutput,
+  ToolResult,
+  ToolResultOptions,
+} from "./type";

packages/core/src/llms/type.ts

@@ -0,0 +1,245 @@
+import type { Tokenizers } from "@llamaindex/env";
+import type { JSONSchemaType } from "ajv";
+import type { JSONObject, JSONValue } from "../global/type";
+
+/**
+ * @internal
+ */
+export interface LLMChat<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+> {
+  chat(
+    params:
+      | LLMChatParamsStreaming<AdditionalChatOptions>
+      | LLMChatParamsNonStreaming<AdditionalChatOptions>,
+  ): Promise<
+    | ChatResponse<AdditionalMessageOptions>
+    | AsyncIterable<ChatResponseChunk<AdditionalMessageOptions>>
+  >;
+}
+
+/**
+ * Unified language model interface
+ */
+export interface LLM<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+> extends LLMChat<AdditionalChatOptions> {
+  metadata: LLMMetadata;
+  /**
+   * Get a chat response from the LLM
+   */
+  chat(
+    params: LLMChatParamsStreaming<
+      AdditionalChatOptions,
+      AdditionalMessageOptions
+    >,
+  ): Promise<AsyncIterable<ChatResponseChunk>>;
+  chat(
+    params: LLMChatParamsNonStreaming<
+      AdditionalChatOptions,
+      AdditionalMessageOptions
+    >,
+  ): Promise<ChatResponse<AdditionalMessageOptions>>;
+
+  /**
+   * Get a prompt completion from the LLM
+   */
+  complete(
+    params: LLMCompletionParamsStreaming,
+  ): Promise<AsyncIterable<CompletionResponse>>;
+  complete(
+    params: LLMCompletionParamsNonStreaming,
+  ): Promise<CompletionResponse>;
+}
+
+export type MessageType = "user" | "assistant" | "system" | "memory";
+
+export type TextChatMessage<AdditionalMessageOptions extends object = object> =
+  {
+    content: string;
+    role: MessageType;
+    options?: undefined | AdditionalMessageOptions;
+  };
+
+export type ChatMessage<AdditionalMessageOptions extends object = object> = {
+  content: MessageContent;
+  role: MessageType;
+  options?: undefined | AdditionalMessageOptions;
+};
+
+export interface ChatResponse<
+  AdditionalMessageOptions extends object = object,
+> {
+  message: ChatMessage<AdditionalMessageOptions>;
+  /**
+   * Raw response from the LLM
+   *
+   * If LLM response an iterable of chunks, this will be an array of those chunks
+   */
+  raw: object | null;
+}
+
+export type ChatResponseChunk<
+  AdditionalMessageOptions extends object = object,
+> = {
+  raw: object | null;
+  delta: string;
+  options?: undefined | AdditionalMessageOptions;
+};
+
+export interface CompletionResponse {
+  text: string;
+  /**
+   * Raw response from the LLM
+   *
+   * It's possible that this is `null` if the LLM response an iterable of chunks
+   */
+  raw: object | null;
+}
+
+export type LLMMetadata = {
+  model: string;
+  temperature: number;
+  topP: number;
+  maxTokens?: number;
+  contextWindow: number;
+  tokenizer: Tokenizers | undefined;
+};
+
+export interface LLMChatParamsBase<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+> {
+  messages: ChatMessage<AdditionalMessageOptions>[];
+  additionalChatOptions?: AdditionalChatOptions;
+  tools?: BaseTool[];
+}
+
+export interface LLMChatParamsStreaming<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+> extends LLMChatParamsBase<AdditionalChatOptions, AdditionalMessageOptions> {
+  stream: true;
+}
+
+export interface LLMChatParamsNonStreaming<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+> extends LLMChatParamsBase<AdditionalChatOptions, AdditionalMessageOptions> {
+  stream?: false;
+}
+
+export interface LLMCompletionParamsBase {
+  prompt: MessageContent;
+}
+
+export interface LLMCompletionParamsStreaming extends LLMCompletionParamsBase {
+  stream: true;
+}
+
+export interface LLMCompletionParamsNonStreaming
+  extends LLMCompletionParamsBase {
+  stream?: false | null;
+}
+
+export type MessageContentTextDetail = {
+  type: "text";
+  text: string;
+};
+
+export type MessageContentImageDetail = {
+  type: "image_url";
+  image_url: { url: string };
+};
+
+export type MessageContentDetail =
+  | MessageContentTextDetail
+  | MessageContentImageDetail;
+
+/**
+ * Extended type for the content of a message that allows for multi-modal messages.
+ */
+export type MessageContent = string | MessageContentDetail[];
+
+export type ToolCall = {
+  name: string;
+  input: JSONObject;
+  id: string;
+};
+
+// happened in streaming response, the tool call is not ready yet
+export type PartialToolCall = {
+  name: string;
+  id: string;
+  // input is not ready yet, JSON.parse(input) will throw an error
+  input: string;
+};
+
+export type ToolResult = {
+  id: string;
+  result: string;
+  isError: boolean;
+};
+
+export type ToolCallOptions = {
+  toolCall: (ToolCall | PartialToolCall)[];
+};
+
+export type ToolResultOptions = {
+  toolResult: ToolResult;
+};
+
+export type ToolCallLLMMessageOptions =
+  | ToolResultOptions
+  | ToolCallOptions
+  | {};
+
+type Known =
+  | { [key: string]: Known }
+  | [Known, ...Known[]]
+  | Known[]
+  | number
+  | string
+  | boolean
+  | null;
+
+export type ToolMetadata<
+  Parameters extends Record<string, unknown> = Record<string, unknown>,
+> = {
+  description: string;
+  name: string;
+  /**
+   * OpenAI uses JSON Schema to describe the parameters that a tool can take.
+   * @link https://json-schema.org/understanding-json-schema
+   */
+  parameters?: Parameters;
+};
+
+/**
+ * Simple Tool interface. Likely to change.
+ */
+export interface BaseTool<Input = any> {
+  /**
+   * This could be undefined if the implementation is not provided,
+   *  which might be the case when communicating with a llm.
+   *
+   * @return {JSONValue | Promise<JSONValue>} The output of the tool.
+   */
+  call?: (input: Input) => JSONValue | Promise<JSONValue>;
+  metadata: // if user input any, we cannot check the schema
+  Input extends Known ? ToolMetadata<JSONSchemaType<Input>> : ToolMetadata;
+}
+
+export type BaseToolWithCall<Input = any> = Omit<BaseTool<Input>, "call"> & {
+  call: NonNullable<Pick<BaseTool<Input>, "call">["call"]>;
+};
+
+export type ToolOutput = {
+  tool: BaseTool | undefined;
+  // all of existing function calling LLMs only support object input
+  input: JSONObject;
+  output: JSONValue;
+  isError: boolean;
+};

packages/core/src/schema/zod.ts

@@ -1,5 +1,7 @@
 import { z } from "zod";
 
+export const anyFunctionSchema = z.function(z.tuple([]).rest(z.any()), z.any());
+
 export const toolMetadataSchema = z.object({
   description: z.string(),
   name: z.string(),
@@ -7,7 +9,7 @@ export const toolMetadataSchema = z.object({
 });
 
 export const baseToolSchema = z.object({
-  call: z.optional(z.function()),
+  call: anyFunctionSchema.optional(),
   metadata: toolMetadataSchema,
 });
 

packages/core/src/utils/event-caller.ts

@@ -1,5 +1,14 @@
 import { AsyncLocalStorage, randomUUID } from "@llamaindex/env";
-import { isAsyncIterable, isIterable } from "../utils.js";
+
+export const isAsyncIterable = (
+  obj: unknown,
+): obj is AsyncIterable<unknown> => {
+  return obj != null && typeof obj === "object" && Symbol.asyncIterator in obj;
+};
+
+export const isIterable = (obj: unknown): obj is Iterable<unknown> => {
+  return obj != null && typeof obj === "object" && Symbol.iterator in obj;
+};
 
 const eventReasonAsyncLocalStorage = new AsyncLocalStorage<EventCaller>();
 

packages/core/src/utils/index.ts

@@ -0,0 +1,54 @@
+export { wrapEventCaller } from "./event-caller";
+
+export async function* streamConverter<S, D>(
+  stream: AsyncIterable<S>,
+  converter: (s: S) => D | null,
+): AsyncIterable<D> {
+  for await (const data of stream) {
+    const newData = converter(data);
+    if (newData === null) {
+      return;
+    }
+    yield newData;
+  }
+}
+
+export async function* streamCallbacks<S>(
+  stream: AsyncIterable<S>,
+  callbacks: {
+    finished?: (value?: S) => void;
+  },
+): AsyncIterable<S> {
+  let value: S | undefined;
+  for await (value of stream) {
+    yield value;
+  }
+  if (callbacks.finished) {
+    callbacks.finished(value);
+  }
+}
+
+export async function* streamReducer<S, D>(params: {
+  stream: AsyncIterable<S>;
+  reducer: (previousValue: D, currentValue: S) => D;
+  initialValue: D;
+  finished?: (value: D) => void;
+}): AsyncIterable<S> {
+  let value = params.initialValue;
+  for await (const data of params.stream) {
+    value = params.reducer(value, data);
+    yield data;
+  }
+  if (params.finished) {
+    params.finished(value);
+  }
+}
+
+export { wrapLLMEvent } from "./wrap-llm-event";
+
+export {
+  extractDataUrlComponents,
+  extractImage,
+  extractSingleText,
+  extractText,
+} from "./llms";

packages/core/src/utils/llms.ts

@@ -0,0 +1,79 @@
+import type {
+  MessageContent,
+  MessageContentDetail,
+  MessageContentTextDetail,
+} from "../llms";
+import type { ImageType } from "../schema";
+
+/**
+ * Extracts just the text from a multi-modal message or the message itself if it's just text.
+ *
+ * @param message The message to extract text from.
+ * @returns The extracted text
+ */
+export function extractText(message: MessageContent): string {
+  if (typeof message !== "string" && !Array.isArray(message)) {
+    console.warn(
+      "extractText called with non-MessageContent message, this is likely a bug.",
+    );
+    return `${message}`;
+  } else if (typeof message !== "string" && Array.isArray(message)) {
+    // message is of type MessageContentDetail[] - retrieve just the text parts and concatenate them
+    // so we can pass them to the context generator
+    return message
+      .filter((c): c is MessageContentTextDetail => c.type === "text")
+      .map((c) => c.text)
+      .join("\n\n");
+  } else {
+    return message;
+  }
+}
+
+/**
+ * Extracts a single text from a multi-modal message content
+ *
+ * @param message The message to extract images from.
+ * @returns The extracted images
+ */
+export function extractSingleText(
+  message: MessageContentDetail,
+): string | null {
+  if (message.type === "text") {
+    return message.text;
+  }
+  return null;
+}
+
+/**
+ * Extracts an image from a multi-modal message content
+ *
+ * @param message The message to extract images from.
+ * @returns The extracted images
+ */
+export function extractImage(message: MessageContentDetail): ImageType | null {
+  if (message.type === "image_url") {
+    return new URL(message.image_url.url);
+  }
+  return null;
+}
+
+export const extractDataUrlComponents = (
+  dataUrl: string,
+): {
+  mimeType: string;
+  base64: string;
+} => {
+  const parts = dataUrl.split(";base64,");
+
+  if (parts.length !== 2 || !parts[0].startsWith("data:")) {
+    throw new Error("Invalid data URL");
+  }
+
+  const mimeType = parts[0].slice(5);
+  const base64 = parts[1];
+
+  return {
+    mimeType,
+    base64,
+  };
+};

packages/core/src/utils/wrap-llm-event.ts

@@ -0,0 +1,80 @@
+import { AsyncLocalStorage, randomUUID } from "@llamaindex/env";
+import { getCallbackManager } from "../global/settings/callback-manager";
+import type { ChatResponse, ChatResponseChunk, LLM, LLMChat } from "../llms";
+
+export function wrapLLMEvent<
+  AdditionalChatOptions extends object = object,
+  AdditionalMessageOptions extends object = object,
+>(
+  originalMethod: LLMChat<
+    AdditionalChatOptions,
+    AdditionalMessageOptions
+  >["chat"],
+  _context: ClassMethodDecoratorContext,
+) {
+  return async function withLLMEvent(
+    this: LLM<AdditionalChatOptions, AdditionalMessageOptions>,
+    ...params: Parameters<
+      LLMChat<AdditionalChatOptions, AdditionalMessageOptions>["chat"]
+    >
+  ): ReturnType<
+    LLMChat<AdditionalChatOptions, AdditionalMessageOptions>["chat"]
+  > {
+    const id = randomUUID();
+    getCallbackManager().dispatchEvent("llm-start", {
+      id,
+      messages: params[0].messages,
+    });
+    const response = await originalMethod.call(this, ...params);
+    if (Symbol.asyncIterator in response) {
+      // save snapshot to restore it after the response is done
+      const snapshot = AsyncLocalStorage.snapshot();
+      const originalAsyncIterator = {
+        [Symbol.asyncIterator]: response[Symbol.asyncIterator].bind(response),
+      };
+      response[Symbol.asyncIterator] = async function* () {
+        const finalResponse = {
+          raw: [] as ChatResponseChunk[],
+          message: {
+            content: "",
+            role: "assistant",
+            options: {},
+          },
+        } satisfies ChatResponse;
+        let firstOne = false;
+        for await (const chunk of originalAsyncIterator) {
+          if (!firstOne) {
+            firstOne = true;
+            finalResponse.message.content = chunk.delta;
+          } else {
+            finalResponse.message.content += chunk.delta;
+          }
+          if (chunk.options) {
+            finalResponse.message.options = {
+              ...finalResponse.message.options,
+              ...chunk.options,
+            };
+          }
+          getCallbackManager().dispatchEvent("llm-stream", {
+            id,
+            chunk,
+          });
+          finalResponse.raw.push(chunk);
+          yield chunk;
+        }
+        snapshot(() => {
+          getCallbackManager().dispatchEvent("llm-end", {
+            id,
+            response: finalResponse,
+          });
+        });
+      };
+    } else {
+      getCallbackManager().dispatchEvent("llm-end", {
+        id,
+        response,
+      });
+    }
+    return response;
+  };
+}

packages/core/tests/event-system.test.ts

@@ -0,0 +1,67 @@
+import { CallbackManager, Settings } from "@llamaindex/core/global";
+import { beforeEach, describe, expect, expectTypeOf, test, vi } from "vitest";
+
+declare module "@llamaindex/core/global" {
+  interface LlamaIndexEventMaps {
+    test: {
+      value: number;
+    };
+  }
+}
+
+describe("event system", () => {
+  beforeEach(() => {
+    Settings.callbackManager = new CallbackManager();
+  });
+
+  test("type system", () => {
+    Settings.callbackManager.on("test", (event) => {
+      const data = event.detail;
+      expectTypeOf(data).not.toBeAny();
+      expectTypeOf(data).toEqualTypeOf<{
+        value: number;
+      }>();
+    });
+  });
+
+  test("dispatch event", async () => {
+    let callback;
+    Settings.callbackManager.on(
+      "test",
+      (callback = vi.fn((event) => {
+        const data = event.detail;
+        expect(data.value).toBe(42);
+      })),
+    );
+
+    Settings.callbackManager.dispatchEvent("test", {
+      value: 42,
+    });
+    expect(callback).toHaveBeenCalledTimes(0);
+    await new Promise((resolve) => process.nextTick(resolve));
+    expect(callback).toHaveBeenCalledTimes(1);
+  });
+
+  // rollup doesn't support decorators for now
+  // test('wrap event caller', async () => {
+  //   class A {
+  //     @wrapEventCaller
+  //     fn() {
+  //       Settings.callbackManager.dispatchEvent('test', {
+  //         value: 42
+  //       });
+  //     }
+  //   }
+  //   const a = new A();
+  //   let callback;
+  //   Settings.callbackManager.on('test', callback = vi.fn((event) => {
+  //     const data = event.detail;
+  //     expect(event.reason!.caller).toBe(a);
+  //     expect(data.value).toBe(42);
+  //   }));
+  //   a.fn();
+  //   expect(callback).toHaveBeenCalledTimes(0);
+  //   await new Promise((resolve) => process.nextTick(resolve));
+  //   expect(callback).toHaveBeenCalledTimes(1);
+  // })
+});

packages/core/tsconfig.json

@@ -8,6 +8,7 @@
     "moduleResolution": "Bundler",
     "skipLibCheck": true,
     "strict": true,
+    "lib": ["ESNext", "DOM"],
     "types": ["node"]
   },
   "include": ["./src"],

packages/env/CHANGELOG.md

@@ -1,5 +1,17 @@
 # @llamaindex/env
 
+## 0.1.8
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+
+## 0.1.7
+
+### Patch Changes
+
+- 41fe871: Add support for azure dynamic session tool
+
 ## 0.1.6
 
 ### Patch Changes

packages/env/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/env",
   "description": "environment wrapper, supports all JS environment including node, deno, bun, edge runtime, and cloudflare worker",
-  "version": "0.1.6",
+  "version": "0.1.8",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/env/src/fs/node.ts

@@ -5,6 +5,7 @@
  *
  * @module
  */
+import { createWriteStream } from "node:fs";
 import fs from "node:fs/promises";
 
-export { fs };
+export { createWriteStream, fs };

packages/env/src/index.ts

@@ -15,12 +15,14 @@ import { ok } from "node:assert";
 import { createHash, randomUUID } from "node:crypto";
 import { EOL } from "node:os";
 import path from "node:path";
+import { Readable } from "node:stream";
 import {
   ReadableStream,
   TransformStream,
   WritableStream,
 } from "node:stream/web";
-import { fs } from "./fs/node.js";
+import { fileURLToPath } from "node:url";
+import { createWriteStream, fs } from "./fs/node.js";
 import type { SHA256 } from "./polyfill.js";
 
 export function createSHA256(): SHA256 {
@@ -38,11 +40,14 @@ export function createSHA256(): SHA256 {
 export { Tokenizers, tokenizers, type Tokenizer } from "./tokenizers/node.js";
 export { AsyncLocalStorage, CustomEvent, getEnv, setEnvs } from "./utils.js";
 export {
+  createWriteStream,
   EOL,
+  fileURLToPath,
   fs,
   ok,
   path,
   randomUUID,
+  Readable,
   ReadableStream,
   TransformStream,
   WritableStream,

packages/experimental/CHANGELOG.md

@@ -1,5 +1,72 @@
 # @llamaindex/experimental
 
+## 0.0.50
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.49
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.48
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.47
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.46
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.45
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.44
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.43
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.42
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.41
 
 ### Patch Changes

packages/experimental/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@llamaindex/experimental",
   "description": "Experimental package for LlamaIndexTS",
-  "version": "0.0.41",
+  "version": "0.0.50",
   "type": "module",
   "types": "dist/type/index.d.ts",
   "main": "dist/cjs/index.js",

packages/llamaindex/CHANGELOG.md

@@ -1,5 +1,89 @@
 # llamaindex
 
+## 0.5.0
+
+### Minor Changes
+
+- 16ef5dd: refactor: simplify callback manager
+
+  Change `event.detail.payload` to `event.detail`
+
+### Patch Changes
+
+- 16ef5dd: refactor: move callback manager & llm to core module
+
+  For people who import `llamaindex/llms/base` or `llamaindex/llms/utils`,
+  use `@llamaindex/core/llms` and `@llamaindex/core/utils` instead.
+
+- 36ddec4: fix: typo in custom page separator parameter for LlamaParse
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - @llamaindex/core@0.1.0
+  - @llamaindex/cloud@0.1.4
+
+## 0.4.14
+
+### Patch Changes
+
+- Updated dependencies [1c444d5]
+  - @llamaindex/cloud@0.1.3
+
+## 0.4.13
+
+### Patch Changes
+
+- e8f8bea: feat: add boundingBox and targetPages to LlamaParseReader
+- 304484b: feat: add ignoreErrors flag to LlamaParseReader
+
+## 0.4.12
+
+### Patch Changes
+
+- f326ab8: chore: bump version
+- Updated dependencies [f326ab8]
+  - @llamaindex/cloud@0.1.2
+  - @llamaindex/core@0.0.3
+  - @llamaindex/env@0.1.8
+
+## 0.4.11
+
+### Patch Changes
+
+- 8bf5b4a: fix: llama parse input spreadsheet
+
+## 0.4.10
+
+### Patch Changes
+
+- 7dce3d2: fix: disable External Filters for Gemini
+
+## 0.4.9
+
+### Patch Changes
+
+- 3a96a48: fix: anthroipic image input
+
+## 0.4.8
+
+### Patch Changes
+
+- 83ebdfb: fix: next.js build error
+
+## 0.4.7
+
+### Patch Changes
+
+- 41fe871: Add support for azure dynamic session tool
+- 321c39d: fix: generate api as class
+- f7f1af0: fix: throw error when no pipeline found
+- Updated dependencies [41fe871]
+- Updated dependencies [f10b41d]
+- Updated dependencies [321c39d]
+  - @llamaindex/env@0.1.7
+  - @llamaindex/core@0.0.2
+  - @llamaindex/cloud@0.1.1
+
 ## 0.4.6
 
 ### Patch Changes
@@ -63,7 +147,7 @@
 
 ### Patch Changes
 
-- 6bc5bdd: feat: add cache disabling, fast mode, do not unroll columns mode and custom page seperator to LlamaParseReader
+- 6bc5bdd: feat: add cache disabling, fast mode, do not unroll columns mode and custom page separator to LlamaParseReader
 - bf25ff6: fix: polyfill for cloudflare worker
 - e6d6576: chore: use `unpdf`
 

packages/llamaindex/e2e/examples/cloudflare-worker-agent/CHANGELOG.md

@@ -1,5 +1,72 @@
 # @llamaindex/cloudflare-worker-agent-test
 
+## 0.0.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.29
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.28
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.27
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.26
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.25
 
 ### Patch Changes

packages/llamaindex/e2e/examples/cloudflare-worker-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/cloudflare-worker-agent-test",
-  "version": "0.0.25",
+  "version": "0.0.34",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/e2e/examples/nextjs-agent/CHANGELOG.md

@@ -1,5 +1,72 @@
 # @llamaindex/next-agent-test
 
+## 0.1.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.1.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.1.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.1.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.1.29
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.1.28
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.1.27
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.1.26
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.1.25
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-agent-test",
-  "version": "0.1.25",
+  "version": "0.1.34",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-agent/src/actions/index.tsx

@@ -1,7 +1,7 @@
 "use server";
 import { createStreamableUI } from "ai/rsc";
+import type { ChatMessage } from "llamaindex";
 import { OpenAIAgent } from "llamaindex";
-import type { ChatMessage } from "llamaindex/llm/types";
 
 export async function chatWithAgent(
   question: string,

packages/llamaindex/e2e/examples/nextjs-edge-runtime/CHANGELOG.md

@@ -1,5 +1,72 @@
 # test-edge-runtime
 
+## 0.1.33
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.1.32
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.1.31
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.1.30
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.1.29
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.1.28
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.1.27
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.1.26
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.1.25
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.1.24
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-edge-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/nextjs-edge-runtime-test",
-  "version": "0.1.24",
+  "version": "0.1.33",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-node-runtime/CHANGELOG.md

@@ -1,5 +1,73 @@
 # @llamaindex/next-node-runtime
 
+## 0.0.15
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.14
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.13
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.12
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.11
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.10
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.9
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.8
+
+### Patch Changes
+
+- 83ebdfb: fix: next.js build error
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.7
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.6
 
 ### Patch Changes

packages/llamaindex/e2e/examples/nextjs-node-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/next-node-runtime-test",
-  "version": "0.0.6",
+  "version": "0.0.15",
   "private": true,
   "scripts": {
     "dev": "next dev",

packages/llamaindex/e2e/examples/nextjs-node-runtime/src/actions/openai.ts

@@ -0,0 +1,68 @@
+"use server";
+import {
+  OpenAI,
+  OpenAIAgent,
+  QueryEngineTool,
+  Settings,
+  VectorStoreIndex,
+} from "llamaindex";
+import { HuggingFaceEmbedding } from "llamaindex/embeddings/HuggingFaceEmbedding";
+import { SimpleDirectoryReader } from "llamaindex/readers/SimpleDirectoryReader";
+
+Settings.llm = new OpenAI({
+  // eslint-disable-next-line turbo/no-undeclared-env-vars
+  apiKey: process.env.NEXT_PUBLIC_OPENAI_KEY ?? "FAKE_KEY_TO_PASS_TESTS",
+  model: "gpt-4o",
+});
+Settings.embedModel = new HuggingFaceEmbedding({
+  modelType: "BAAI/bge-small-en-v1.5",
+  quantized: false,
+});
+Settings.callbackManager.on("llm-tool-call", (event) => {
+  console.log(event.detail);
+});
+Settings.callbackManager.on("llm-tool-result", (event) => {
+  console.log(event.detail);
+});
+
+export async function getOpenAIModelRequest(query: string) {
+  try {
+    const currentDir = __dirname;
+
+    // load our data and create a query engine
+    const reader = new SimpleDirectoryReader();
+    const documents = await reader.loadData(currentDir);
+    const index = await VectorStoreIndex.fromDocuments(documents);
+    const retriever = index.asRetriever({
+      similarityTopK: 10,
+    });
+    const queryEngine = index.asQueryEngine({
+      retriever,
+    });
+
+    // define the query engine as a tool
+    const tools = [
+      new QueryEngineTool({
+        queryEngine: queryEngine,
+        metadata: {
+          name: "deployment_details_per_env",
+          description: `This tool can answer detailed questions about deployments happened in various environments.`,
+        },
+      }),
+    ];
+    // create the agent
+    const agent = new OpenAIAgent({ tools });
+
+    const { response } = await agent.chat({
+      message: query,
+    });
+    return {
+      message: response,
+    };
+  } catch (err) {
+    console.error(err);
+    return {
+      errors: "Error Calling OpenAI Model",
+    };
+  }
+}

packages/llamaindex/e2e/examples/nextjs-node-runtime/src/app/api/openai/route.ts

@@ -0,0 +1,10 @@
+import { getOpenAIModelRequest } from "@/actions/openai";
+import { NextRequest, NextResponse } from "next/server";
+
+// POST /api/openai
+export async function POST(request: NextRequest) {
+  const body = await request.json();
+  const content = await getOpenAIModelRequest(body.query);
+
+  return NextResponse.json(content, { status: 200 });
+}

packages/llamaindex/e2e/examples/waku-query-engine/CHANGELOG.md

@@ -1,5 +1,72 @@
 # @llamaindex/waku-query-engine-test
 
+## 0.0.34
+
+### Patch Changes
+
+- Updated dependencies [16ef5dd]
+- Updated dependencies [16ef5dd]
+- Updated dependencies [36ddec4]
+  - llamaindex@0.5.0
+
+## 0.0.33
+
+### Patch Changes
+
+- llamaindex@0.4.14
+
+## 0.0.32
+
+### Patch Changes
+
+- Updated dependencies [e8f8bea]
+- Updated dependencies [304484b]
+  - llamaindex@0.4.13
+
+## 0.0.31
+
+### Patch Changes
+
+- Updated dependencies [f326ab8]
+  - llamaindex@0.4.12
+
+## 0.0.30
+
+### Patch Changes
+
+- Updated dependencies [8bf5b4a]
+  - llamaindex@0.4.11
+
+## 0.0.29
+
+### Patch Changes
+
+- Updated dependencies [7dce3d2]
+  - llamaindex@0.4.10
+
+## 0.0.28
+
+### Patch Changes
+
+- Updated dependencies [3a96a48]
+  - llamaindex@0.4.9
+
+## 0.0.27
+
+### Patch Changes
+
+- Updated dependencies [83ebdfb]
+  - llamaindex@0.4.8
+
+## 0.0.26
+
+### Patch Changes
+
+- Updated dependencies [41fe871]
+- Updated dependencies [321c39d]
+- Updated dependencies [f7f1af0]
+  - llamaindex@0.4.7
+
 ## 0.0.25
 
 ### Patch Changes

packages/llamaindex/e2e/examples/waku-query-engine/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@llamaindex/waku-query-engine-test",
-  "version": "0.0.25",
+  "version": "0.0.34",
   "type": "module",
   "private": true,
   "scripts": {

packages/llamaindex/e2e/fixtures/llm/openai.ts

@@ -1,3 +1,4 @@
+import { extractText } from "@llamaindex/core/utils";
 import type {
   ChatResponse,
   ChatResponseChunk,
@@ -7,8 +8,7 @@ import type {
   LLMChatParamsStreaming,
   LLMCompletionParamsNonStreaming,
   LLMCompletionParamsStreaming,
-} from "llamaindex/llm/types";
-import { extractText } from "llamaindex/llm/utils";
+} from "llamaindex";
 import { deepStrictEqual, strictEqual } from "node:assert";
 import { llmCompleteMockStorage } from "../../node/utils.js";
 

packages/llamaindex/e2e/node/claude.e2e.ts

@@ -1,7 +1,7 @@
+import { extractText } from "@llamaindex/core/utils";
 import { consola } from "consola";
 import { Anthropic, FunctionTool, Settings, type LLM } from "llamaindex";
 import { AnthropicAgent } from "llamaindex/agent/anthropic";
-import { extractText } from "llamaindex/llm/utils";
 import { ok } from "node:assert";
 import { beforeEach, test } from "node:test";
 import { getWeatherTool, sumNumbersTool } from "./fixtures/tools.js";

packages/llamaindex/e2e/node/openai.e2e.ts

@@ -1,3 +1,4 @@
+import { extractText } from "@llamaindex/core/utils";
 import { consola } from "consola";
 import {
   Document,
@@ -14,7 +15,6 @@ import {
   VectorStoreIndex,
   type LLM,
 } from "llamaindex";
-import { extractText } from "llamaindex/llm/utils";
 import { ok, strictEqual } from "node:assert";
 import { readFile } from "node:fs/promises";
 import { join } from "node:path";

packages/llamaindex/e2e/node/react.e2e.ts

@@ -1,5 +1,5 @@
+import { extractText } from "@llamaindex/core/utils";
 import { OpenAI, ReActAgent, Settings, type LLM } from "llamaindex";
-import { extractText } from "llamaindex/llm/utils";
 import { ok } from "node:assert";
 import { beforeEach, test } from "node:test";
 import { getWeatherTool } from "./fixtures/tools.js";

packages/llamaindex/e2e/node/utils.ts

@@ -4,16 +4,17 @@ import {
   type LLMEndEvent,
   type LLMStartEvent,
   type LLMStreamEvent,
-} from "llamaindex";
+} from "@llamaindex/core/global";
+import { CustomEvent } from "@llamaindex/env";
 import { readFile, writeFile } from "node:fs/promises";
 import { join } from "node:path";
 import { type test } from "node:test";
 import { fileURLToPath } from "node:url";
 
 type MockStorage = {
-  llmEventStart: LLMStartEvent["detail"]["payload"][];
-  llmEventEnd: LLMEndEvent["detail"]["payload"][];
-  llmEventStream: LLMStreamEvent["detail"]["payload"][];
+  llmEventStart: LLMStartEvent[];
+  llmEventEnd: LLMEndEvent[];
+  llmEventStream: LLMStreamEvent[];
 };
 
 export const llmCompleteMockStorage: MockStorage = {
@@ -36,35 +37,35 @@ export async function mockLLMEvent(
     llmEventStream: [],
   };
 
-  function captureLLMStart(event: LLMStartEvent) {
-    idMap.set(event.detail.payload.id, `PRESERVE_${counter++}`);
+  function captureLLMStart(event: CustomEvent<LLMStartEvent>) {
+    idMap.set(event.detail.id, `PRESERVE_${counter++}`);
     newLLMCompleteMockStorage.llmEventStart.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
     });
   }
 
-  function captureLLMEnd(event: LLMEndEvent) {
+  function captureLLMEnd(event: CustomEvent<LLMEndEvent>) {
     newLLMCompleteMockStorage.llmEventEnd.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
       response: {
-        ...event.detail.payload.response,
+        ...event.detail.response,
         // hide raw object since it might too big
         raw: null,
       },
     });
   }
 
-  function captureLLMStream(event: LLMStreamEvent) {
+  function captureLLMStream(event: CustomEvent<LLMStreamEvent>) {
     newLLMCompleteMockStorage.llmEventStream.push({
-      ...event.detail.payload,
+      ...event.detail,
       // @ts-expect-error id is not UUID, but it is fine for testing
       id: idMap.get(event.detail.payload.id)!,
       chunk: {
-        ...event.detail.payload.chunk,
+        ...event.detail.chunk,
         // hide raw object since it might too big
         raw: null,
       },

packages/llamaindex/e2e/package.json

@@ -10,6 +10,7 @@
   },
   "devDependencies": {
     "@faker-js/faker": "^8.4.1",
+    "@llamaindex/core": "workspace:*",
     "@types/node": "^20.12.11",
     "consola": "^3.2.3",
     "llamaindex": "workspace:*",

packages/llamaindex/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llamaindex",
-  "version": "0.4.6",
+  "version": "0.5.0",
   "license": "MIT",
   "type": "module",
   "keywords": [
@@ -22,6 +22,7 @@
   "dependencies": {
     "@anthropic-ai/sdk": "^0.21.1",
     "@aws-crypto/sha256-js": "^5.2.0",
+    "@azure/identity": "^4.2.1",
     "@datastax/astra-db-ts": "^1.2.1",
     "@google-cloud/vertexai": "^1.2.0",
     "@google/generative-ai": "^0.12.0",
@@ -33,7 +34,8 @@
     "@mistralai/mistralai": "^0.4.0",
     "@pinecone-database/pinecone": "^2.2.2",
     "@qdrant/js-client-rest": "^1.9.0",
-    "@types/lodash": "^4.17.5",
+    "@types/lodash": "^4.17.4",
+    "@types/node": "^20.14.5",
     "@types/papaparse": "^5.3.14",
     "@types/pg": "^8.11.6",
     "@xenova/transformers": "^2.17.2",

packages/llamaindex/src/ChatHistory.ts

@@ -1,9 +1,9 @@
+import type { ChatMessage, LLM, MessageType } from "@llamaindex/core/llms";
+import { extractText } from "@llamaindex/core/utils";
 import { tokenizers, type Tokenizer } from "@llamaindex/env";
 import type { SummaryPrompt } from "./Prompt.js";
 import { defaultSummaryPrompt, messagesToHistoryStr } from "./Prompt.js";
 import { OpenAI } from "./llm/openai.js";
-import type { ChatMessage, LLM, MessageType } from "./llm/types.js";
-import { extractText } from "./llm/utils.js";
 
 /**
  * A ChatHistory is used to keep the state of back and forth chat messages

packages/llamaindex/src/EngineResponse.ts

@@ -1,10 +1,10 @@
-import type { NodeWithScore } from "@llamaindex/core/schema";
 import type {
   ChatMessage,
   ChatResponse,
   ChatResponseChunk,
-} from "./llm/types.js";
-import { extractText } from "./llm/utils.js";
+} from "@llamaindex/core/llms";
+import type { NodeWithScore } from "@llamaindex/core/schema";
+import { extractText } from "@llamaindex/core/utils";
 
 export class EngineResponse implements ChatResponse, ChatResponseChunk {
   sourceNodes?: NodeWithScore[];